Transportation – like all industries – is undergoing a tech-led transformation.
And now MIT researchers have developed software that uses sound and vibration to alert people of impending problems in cars and trucks. Harnessing machine learning, the sounds and movement of a vehicle would send information, via a smartphone app, to the user. By picking up telltale diagnostic signals, the smartphone’s ‘microphone and accelerometers’, without any prior knowledge of the vehicle – and with no need to be ‘connected’ to it digitally, detects a problem that needs addressing. This could relate to air pressure in a tire, engine problems or other common issues.
The researchers are currently working on a prototype smartphone app with the range of diagnostic indicators it has developed – which should be ready by spring 2018.
“For our data, we’ve induced failures [in] a perfectly good vehicle” fixed it and then “returned the car better than when we took it out. I’ve rented cars and given them new air filters, balanced their tires, and done an oil change,” explained research scientist Joshua Siegel.